Loading...

Single-Channel Speech Dereverberation in Noisy Acoustical Environments

Joorabchi, Marjan | 2015

638 Viewed
  1. Type of Document: M.Sc. Thesis
  2. Language: English
  3. Document No: 47148 (56)
  4. University: Sharif University of Technology, International Campus, Kish Island
  5. Department: Science and Engineering
  6. Advisor(s): Ghorshi, Mohammad Ali
  7. Abstract:
  8. In speech processing, reflections of sound wave in a bounded space are considered as speech reverberation. Although for musical instruments and their related recording devices these reflections are useful, however, some other applications face serious problems receiving them along with speech signal. Reverberation causes speech degradation and intelligibility as well as highly quality reduction. Dereverberation algorithms are essential for Automatic Speech Recognition (ASR), telecommunication and hearing aid devices, which are some of the mostly used applications. While dereverberation itself is a challenging problem, dereverberate a speech signal recorded only by one microphone (channel) requires more attention and resolution due to obtain and analyze one observed speech signal. Aside from that, in noisy acoustical environments interference signals such as white noise decrease quality of received signal even more. In this thesis, we produce a reverberated speech signal, add noise to it, and then dereverberate it in a way the processed speech signal contains no noise and less reverberation. Adaptive filters used in this research are modified and optimized in order to solve reverberation and noise problems. Also, two different proposed scenarios are used in this thesis with different Signal-to-Noise ratios of Additive White Gaussian Noise used in these experiments and the results are compared together. The experimental results of these methods show significant improvement after a noisy-reverberated speech signal has been dereverberated
  9. Keywords:
  10. Speech Signal ; Reverberation ; Adaptive Filters ; Noisy Acoustical Environment ; Dereverberation ; Single-Channel Speech Signal

 Digital Object List

 Bookmark

No TOC